From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33887) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Um4gx-00039j-8R for qemu-devel@nongnu.org; Mon, 10 Jun 2013 12:11:37 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Um4gv-0002LX-Hf for qemu-devel@nongnu.org; Mon, 10 Jun 2013 12:11:35 -0400 Received: from mail-ob0-x229.google.com ([2607:f8b0:4003:c01::229]:43354) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Um4gv-0002LJ-Cj for qemu-devel@nongnu.org; Mon, 10 Jun 2013 12:11:33 -0400 Received: by mail-ob0-f169.google.com with SMTP id up14so10449640obb.14 for ; Mon, 10 Jun 2013 09:11:32 -0700 (PDT) Sender: fluxion Date: Mon, 10 Jun 2013 11:10:29 -0500 From: mdroth Message-ID: <20130610161029.GB5727@vm> References: <1370859260-8183-1-git-send-email-pl@kamp.de> <1370859260-8183-3-git-send-email-pl@kamp.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1370859260-8183-3-git-send-email-pl@kamp.de> Subject: Re: [Qemu-devel] [PATCHv2 2/2] migration: do not overwrite zero pages List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Lieven Cc: quintela@redhat.com, aik@ozlabs.ru, qemu-devel@nongnu.org, owasserm@redhat.com, pbonzini@redhat.com, xiawenc@linux.vnet.ibm.com, david@gibson.dropbear.id.au On Mon, Jun 10, 2013 at 12:14:20PM +0200, Peter Lieven wrote: > on incoming migration do not memset pages to zero if they already read as zero. > this will allocate a new zero page and consume memory unnecessarily. even > if we madvise a MADV_DONTNEED later this will only deallocate the memory > asynchronously. > > Signed-off-by: Peter Lieven > --- > arch_init.c | 14 ++++++++------ > 1 file changed, 8 insertions(+), 6 deletions(-) > > diff --git a/arch_init.c b/arch_init.c > index 08fccf6..cf4e1d5 100644 > --- a/arch_init.c > +++ b/arch_init.c > @@ -832,14 +832,16 @@ static int ram_load(QEMUFile *f, void *opaque, int version_id) > } > > ch = qemu_get_byte(f); > - memset(host, ch, TARGET_PAGE_SIZE); > + if (ch != 0 || !is_zero_page(host)) { > + memset(host, ch, TARGET_PAGE_SIZE); > #ifndef _WIN32 > - if (ch == 0 && > - (!kvm_enabled() || kvm_has_sync_mmu()) && > - getpagesize() <= TARGET_PAGE_SIZE) { > - qemu_madvise(host, TARGET_PAGE_SIZE, QEMU_MADV_DONTNEED); > - } > + if (ch == 0 && > + (!kvm_enabled() || kvm_has_sync_mmu()) && > + getpagesize() <= TARGET_PAGE_SIZE) { > + qemu_madvise(host, TARGET_PAGE_SIZE, QEMU_MADV_DONTNEED); > + } Reviewed-by: Michael Roth Also CC'ing qemu-stable, but from what I gather this just mitigates the performance impact of 1/2 and isn't strictly necessary to fix migration? i.e. we can still do outgoing migration to 1.5.0 guests? > #endif > + } > } else if (flags & RAM_SAVE_FLAG_PAGE) { > void *host; > > -- > 1.7.9.5 > >